Enhancing Autonomous Driving Navigation Using Soft Actor-Critic

Elallid, B. B., Benamar, N., Bagaa, M. et Hadjadj-Aoul, Y. (2024). Enhancing Autonomous Driving Navigation Using Soft Actor-Critic. Future Internet, 16 (7). p. 238. ISSN 1999-5903 DOI 10.3390/fi16070238

Prévisualisation

PDF Disponible sous licence Creative Commons Attribution. Télécharger (4MB) | Prévisualisation

Résumé

Autonomous vehicles have gained extensive attention in recent years, both in academia and industry. For these self-driving vehicles, decision-making in urban environments poses significant challenges due to the unpredictable behavior of traffic participants and intricate road layouts. While existing decision-making approaches based on Deep Reinforcement Learning (DRL) show potential for tackling urban driving situations, they suffer from slow convergence, especially in complex scenarios with high mobility. In this paper, we present a new approach based on the Soft Actor-Critic (SAC) algorithm to control the autonomous vehicle to enter roundabouts smoothly and safely and ensure it reaches its destination without delay. For this, we introduce a destination vector concatenated with extracted features using Convolutional Neural Networks (CNN). To evaluate the performance of our model, we conducted extensive experiments in the CARLA simulator and compared it with the Deep Q-Network (DQN) and Proximal Policy Optimization (PPO) models. Qualitative results reveal that our model converges rapidly and achieves a high success rate in scenarios with high traffic compared to the DQN and PPO models.

Type de document:	Article
Mots-clés libres:	Autonomous driving Deep reinforcement learning Navigation
Date de dépôt:	18 août 2025 15:20
Dernière modification:	18 août 2025 15:20
Version du document déposé:	Version officielle de l'éditeur
URI:	https://depot-e.uqtr.ca/id/eprint/12185

Actions (administrateurs uniquement)

Éditer la notice